Skip to contents

Execute an Expression on a DuckDB Table with Optional Materialization

Source: R/run_mod_dplyr.R
run_mod_dplyr.Rd

Connects to a DuckDB database, evaluates a user-supplied expression on a specified table, and either collects the result into R or computes and stores it as a new table in the database.

Usage

run_mod_dplyr(
  mod_db,
  table_name,
  expr,
  mode = c("collect", "compute"),
  output_table = NULL
)

Arguments

mod_db

Path to the DuckDB database file (e.g., `"my_data.mod.db"`).

table_name

Name of the table in the database to operate on.

expr

A function taking one argument (`tbl_ref`) and returning a lazy `dplyr` expression. The table reference (`tbl_ref`) will be passed as a `tbl()` object connected to the database.

mode

One of `"collect"` (default) or `"compute"`. If `"collect"`, the result is returned as a data frame in R. If `"compute"`, the result is stored as a new table in the database.

output_table

Required if `mode = "compute"`. Name of the output table to create or overwrite with the result of `expr(tbl_ref)`.

Value

If `mode = "collect"`, returns a data frame. If `mode = "compute"`, returns `NULL` invisibly.

Examples

if (FALSE) { # \dontrun{
# Collect results of a filtered table into R
run_mod_dplyr(
  mod_db = "my_data.mod.db",
  table_name = "methylation_data",
  expr = function(tbl_ref) dplyr::filter(tbl_ref, score > 0.5),
  mode = "collect"
)

# Store the filtered result in a new table inside the database
run_mod_dplyr(
  mod_db = "my_data.mod.db",
  table_name = "methylation_data",
  expr = function(tbl_ref) dplyr::filter(tbl_ref, score > 0.5),
  mode = "compute",
  output_table = "filtered_data"
)
} # }